This study is designed to examine in-session therapeutic processes that contribute to engagement and retention in family therapy with drug abusing adolescents. In particular, this study examines common therapist interventions and core family processes that predict failure to engage (i.e., family drop outs prior to sessions 4) and failure to retain (i.e., family drop outs after session 4-7) in treatment. It is hypothesize that for families that engage in therapy, therapist Support and Cognitive Restructuring interventions will lead to increased family Alliance and decreased Conflict/Negativity respectively. It is further hypothesized that for families that retain in therapy, therapist Directive interventions will lead to increases in family Parenting. Families will be selected from the archives of three established family therapy programs (Functional Family Therapy; Multidimensional Family Therapy; Structural; Family Therapy). Common therapist interventions and core family processes will be identified from codings (microsession) and ratings (macrosession) of videotaped sessions. Preliminary confirmatory factor analyses will be conduced to examine the factor, structure of the common therapist interventions and core family processes. HLM techniques will be used to examine the impact of therapist interventions on changes in family processes, and the relationship between these changes and engagement and retention in therapy.